Protonation and Chloroplast Membrane Structure

PROTONATION AND CHLOROPLAST MEMBRANE STRUCTURE

SATORU MURAKAMI and LESTER PACKER

From the Department of Physiology-Anatomy, University of California, Berkeley, California 94720 Downloaded from http://rupress.org/jcb/article-pdf/47/2/332/1069424/332.pdf by guest on 29 September 2021

ABSTRACT Light changes the structure of chloroplasts . This effect was investigated by high resolution electron microscopy, photometric methods, and chemical modification . (a) A reversible contraction of chloroplast membrane occurs upon illumination, dark titration with H+, or increasing osmolarity . These gross structural changes arise from a flattening of the thylakoids, with a corresponding decrease in the spacing between membranes . Microdensitometry showed that illumination or dark addition of H+ resulted in a 13-23% decrease in membrane thickness. Osmotically contracted chloroplasts do not show this effect . (b) Rapid glutaraldehyde fixation during actual experiments revealed that transmission changes are closely correlated with the spacing changes and therefore reflect an osmotic mechanism, whereas the light scattering changes have kinetics most similar to changes in membrane thickness or conformation . (c) Kinetic analysis of light scattering and transmission changes with the changes in fluorescence of anilinonaphthalene sulfonic acid bound to membranes revealed that fluorescence preceded light scattering or transmission changes. (d) It is concluded that the temporal sequence of events following illumination probably are protonation, changes in the environment within the membrane, change in membrane thickness, change in internal osmolarity accompanying ion movements with consequent collapse and flattening of thylakoid, change in the gross morphology of the inner chloroplast membrane system, and change in the gross morphology of whole chloroplasts .

INTRODUCTION

Ample evidence has been brought forward in changes in chloroplast membranes is still undeter- recent years to demonstrate, both in vitro (1-9) mined . and in vivo (10-13), the conditions under which This investigation was undertaken to elucidate light-induced proton and ion transport across the the sequence of steps by which changes in hydrogen chloroplast membrane leads to reversible changes gradients alter the gross morphology of the chloro- in the volume and configuration of these organel- plast membrane system . Our preliminary investi- les. Dilley and Vernon (2), Deamer, Crofts, and gations (14) have revealed that changes in the Packer (4), and Crofts, Deamer, and Packer (5) thickness of the thylakoid membrane are caused have suggested that changes in distribution of by the action of light and accompany the estab- hydrogen ions generated by light reactions of lishment of hydrogen gradients . This contraction chloroplasts play a central role in controlling the of the membrane is distinct from the more com- morphological changes following illumination, monly recognized "flattening" of the chloroplasts . although the precise mechanism whereby changes In order to clarify the nature of these microstruc- in hydrogen ions bring about the morphological tural changes, we initiated the present studies

332 THE JOURNAL OF CELL BIOLOGY . VOLUME 47, 1970 • pages 332-351

using such techniques as high resolution elec- pH Studies tron microscopy, microdensitometry, light-scat- For pH studies, chloroplasts were suspended in tering and transmission measurements, and 100 mm NaCl solution at pH 7 .8 (adjusted with 0.05 chemical probes, i.e., hydrophobic-seeking fluo- N NaOH) and then titrated with 0 .05 N HCI. Changes rochrome (8-anilinonaphthalene-l-sulfonic acid in pH of the incubation were measured with a glass [ANS]) and a sulfhydryl-seeking probe (phen- electrode inserted into the cuvette and connected to a ylmercuric acetate [PMA]) . Radiometer pH meter 22, the output of which was continuously recorded. MATERIALS AND METHODS PMA Binding Preparation of Chloroplasts Chloroplasts (13-15 µg chlorophyll /nil) were Spinach (Spinacia oleracea) leaves were homogenized in 50 mm Tris-HCl buffer, pH 8.0, containing 175 incubated for 4 min in 7 .5 ml of 50 mm Tris-HCI, mm NaCl . The homogenate was filtered through pH 8.0, containing 175 mm NaCl, 15 µM PMS, four layers of cheesecloth and centrifuged for 2 min and 20 µM PMA in the dark, in the light, and in the at 500 g to remove debris . The supernatant was dark following 5 min illumination. Neutral glutaral- Downloaded from http://rupress.org/jcb/article-pdf/47/2/332/1069424/332.pdf by guest on 29 September 2021 decanted and centrifuged again at 1500 g for 4 min . dehyde (N2 scaled, Polyscience Inc., Warrington, The chloroplast fraction was then washed once in an Pa.) was added at a concentration of 0 .1 M, and isolation medium by centrifugation at 1500 g for after 10 min the suspension was removed from the 6 min, resuspended in a small amount of the same cuvette and centrifuged for 10 min at 20,000 g. medium, and stored in an ice bath in the dark until The amount of unreacted PMA in the supernatant used in experiments performed at 25'C . was then assayed by dithizone (6 µg per ml chloro- For the study of photoshrinkage, isolated chloro- form) according to the procedure described by plasts were incubated in either sodium chloride- Miller et al . (15) from readings at 620 nm, and PMA medium (50 mm Tris-HCI, pH 8 .0, 175 mm the amount of PMA taken up by the chloroplast NaCl, 15 AM phenazine methosulfate [PMS] and membranes was calculated by difference from con- 20 µM PMA) or sodium acetate (150 mm sodium trols. acetate, pH 6 .7, plus 15 µM PMS) in the dark and then illuminated by red light. For osmotic studies, ANS Fluorescence chloroplasts were incubated in the dark in either 0.05 or 0 .5 M sucrose solution containing 44 mm To avoid artifacts from actinic light, the difference NaCl and 12 .5 mM Tris-HCI, pH 7 .9. in ANS fluorescence between unfixed and fixed chloroplasts was taken as a measure of the light- 90° Light Scattering and induced structural change. Chloroplasts were treated for 10 min with 0 .1 M glutaraldehyde, and then Transmission Studies washed three times by centrifuging for 10 min at Experiments were carried out in a cuvette in a 20,000 g . Chloroplasts thus "fixed" do not exhibit Brice-Phoenix light-scattering photometer (Phoenix any light scattering and transmission responses Precision Instrument Co., Philadelphia, Pa.) con- upon illumination . The reaction mixture contained nected to a dual channel recorder (recti/riter, Texas chloroplasts, 100 µg ANS and 15 µM PMS in the Instruments, Inc ., Houston, Texas) . The reaction dark, the fluorescence level was adjusted to read mixture was continuously stirred by a small magnetic 100%. Then the reaction mixture was illuminated bar . Light-induced changes in 90 ° light scattering or titrated with HCI in the dark . For experiments and transmission were recorded simultaneously . in a sodium chloride-PMA medium, PMA (20 To prevent a concentration-dependent interference gm) was added before illumination . Identical pro- by multiscattering of light by pigmented particles, cedures were used with fixed chloroplasts. Fluo- the chlorophyll concentration was kept less than 15 rescence was detected between 455 and 578 nm at µg per ml incubation medium. Within this range the low angle (20 ° ) to the exciting beam (290-404 nm) responses are proportional to concentration . The to prevent artifacts from scattering samples . measuring light was filtered at 546 nm with an interference filter, and the intensity in the dark before Electron Microscopy illumination or acidification of the suspension was adjusted to read 100% on the chart paper . Chloro- After 0.1 M glutaraldehyde treatment in the cuvette, plasts were illuminated by red light (600-700 run, reaction mixtures were transferred to test tubes and 900 ft-c) from the side of the cuvette opposite to the fixation was continued for 1 hr under continuous light-scattering measurement . light or dark. Samples were then treated with 1 %

SATORU MtTRAKAMI ANn LESTER PACKER Protonation and Chloroplast Membrane Structure 3 3 3

osmium tetroxide in 50 mm phosphate buffer, pH electron microscope negatives (X 20,000) by micro- 7.4, and then washed with this buffer, dehydrated beam on the double-beam Microdensitometer MK with ethanol, and embedded in Epon-Araldite IIIC (Joyce Loeble & Co., Inc ., Burlington, Mass.) mixture (16) as modified by Johnson and Porter and recorded on the chart paper at 50 times ex- (17) . Sections were observed in an EMU-3H (RCA) pansion . The size of the microbeam used was about or Elmiskop I (Siemens) electron microscope with- 50 p., which is small enough to resolve the thickness out poststaining by heavy metal compounds. and spacing of the membranes even when they become tightly packed, because (a) the width of Dimension of Membranes the grana on the electron microscope negative is 200-300 p. and (b) the interspace or clear zone Thickness and spacing of membranes were esti- between two adjacent membranes is more than 80 µ mated by two methods : (a) direct measurement on at 50 times expansion . Measurement of half-width enlarged photographs at a final magnification of of the peak was done by extrapolating the slope of 200,000 by using a magnifier equipped with micro- the tracing to the background level as shown in scale and (b) microdensitometry tracing of electron Figs . 6, 10, 12, and 15. microscope negatives . In order to obtain measurements closest to real dimensions, certain precautions RESULTS Downloaded from http://rupress.org/jcb/article-pdf/47/2/332/1069424/332.pdf by guest on 29 September 2021 were taken . The grana portion of the membrane system was selected for measurement, since the Kinetics of Light Scattering and grana membrane is extremely flat, and is stacked Transmission Changes with membranes at constant spacing. Under such circumstances, the orientation of the membrane Photometric methods permit evaluation of the with respect to the direction of electron beam can dynamic aspects of light-induced changes in . Since this is very critical, especially for be defined chloroplast structure . Fig. 1 shows that when tightly packed grana structure, stacked membranes chloroplasts are suspended in sodium acetate separated by 30-40 A clear zone were examined . If the orientation of the membrane within the section medium, reversible photoshrinkage of chloroplasts tilted by more than a few degrees, it would not be can be readily monitored by reversible changes in possible to clearly resolve individual membranes . either light scattering or transmission. Such Densitometric curves were obtained by tracings on changes have been observed previously (5) .

180

ç m 160 o Hu -`c 140 01

o O U 120

100

C 100 o EN

1J, 80 C o H 60 1 1 1 1 1 1 1 i 1 1 1 1 1 0 2 4 6 8 10 12 Minutes

FIGURE 1 Kinetics of light-induced 90° light scattering and transmission changes of isolated spinach chloroplasts accompanying photoshrinkage in a sodium acetate medium. Other conditions as in Methods .

3 3 4 THE JOURNAL OF CELL BIOLOGY . VOLUME 47, 1970

100 •

100µg chlorophyll

•

o 1 o' 6 8 10 12 14 16 PMA (heM) FIGURE û Effect of PMA concentration on the light scattering response of illuminated chloroplasts . Downloaded from http://rupress.org/jcb/article-pdf/47/2/332/1069424/332.pdf by guest on 29 September 2021 Chloroplasts suspended in 7 .0 ml of 50 mm Tris-HCI, pH 8 .0, containing 175 mat NaCl and 15 'Um PMS were illuminated by actinic red light and, after 1 min, PMA was added to the incubation at varied concentrations .

TABLE I PMA Uptake by Chloroplast Membranes

PMA bound

Experiments

Incubation 1 2 3 Average

ms moles/mg chlorophyll Dark, 4 min (before illumination) 478 478 461 472 Light, 4 min 494 537 566 532 Dark, 4 min (after 4 min illumination) 500 516 500 505 Chloroplasts (140 µg chlorophyll) were incubated in a sodium chloride-PMA medium . PMA was assayed as described in Methods. The difference between the amount of PMA added initially and the amount detected in the supernatant was taken as the amount of PMA bound . Values are for experiments with chloroplasts from same preparation .

Light-induced chloroplast shrinkage is enhanced Light scattering and transmission changes were by organic mercurial PMA, which has been examined simultaneously in chloroplasts sus- studied by Siegenthaler and Packer (18) and pended in the sodium chloride-PMA medium and Siegenthaler (19). It provides a useful system for were found to be parallel to one another after examining light-induced changes in chloroplast illumination . When the dithiol dithioerythritol structure . Fig . 2 shows the relation between PMA (DTE) is added to chloroplasts in an amount concentration and the magnitude of the light- which exceeds, on a stoichiometric basis (see Table induced light scattering increments observed in II and Fig. 3), the quantity of PMA present, then chloroplasts suspended in sodium chloride-PMA the light-induced changes are reversed, even under medium . Maximal photoshrinkage is induced by the conditions of continuous illumination . Changes 10 Am PMA or less, under usual conditions of occur immediately upon the addition of DTE in chlorophyll concentration in in vitro experiments . both light scattering (decrease) and transmission The enhanced photoshrinkage can be correlated (increase) . However, light scattering decrease with the amount of membrane-bound PMA as reaches a steady state but the transmission change seen in Table I . Upon illumination, an increase continues. This difference in kinetics of light- in the extent of PMA binding occurs which falls scattering and transmission is also seen upon ad- by about 50% when the light is removed . dition of PMA to chloroplasts under continued

SATORU MURAKAMI AND LESTER PACKER Protonation and Chloroplast Membrane Structure 335

TABLE II Reversal of Light-Induced Light Scattering and Transmission Changes of Chloroplasts by DTE

Concentration 90 ° Light scattering Transmission

Inhibition by Inhibition by PMA DTE Increase t''2 DTE Decrease tt/2 DTE

µ:v MM % sec % sec

0 0 12 3 0 20 11 3 10 0 84 48 21 40 10 2 .5 84 60 None 22 45 None 10 5 10 Complete 3 Complete 20 0 103 40 25 35 20 5 102 55 None 24 45 None

20 10 11 Complete 4 Complete Downloaded from http://rupress.org/jcb/article-pdf/47/2/332/1069424/332.pdf by guest on 29 September 2021 20 20 13 Complete 4 Complete 40 10 95 None 24 None 40 20 10 Complete 3 Complete

Chloroplasts were incubated in 50 mm Tris-HC1, pH 8 .0, containing 175 mm NaCl and 15 MM PMS in the presence of PMA and DTE at the concentrations indicated . Illumi- nation was by red light (600-700 nm). Chlorophyll concentration was 120 /Ig in 7 .5 ml of reaction mixture . Half-maximum time (t1 12 ) is presented to show the effect of DTE on the kinetics of both light scattering and transmission changes .

DTE 5 µMl 1 5µM

FIGURE 3 Effect of DTE on light-induced 90° light scattering and transmission changes of chloroplasts . Conditions : sodium chloride-PMA medium ; at the points indicated, 10 µM DTE and 20 µM PMA were added to the incubation during illumination .

3 3 6 THE JOURNAL OF CELL BIOLOGY . VOLUME 47, 1970 Downloaded from http://rupress.org/jcb/article-pdf/47/2/332/1069424/332.pdf by guest on 29 September 2021

FIGURE 4 Ultrastructure of chloroplasts suspended in a sodium chloride-PMA medium under different light conditions . Chloroplasts were incubated and fixed as described in Methods . (A) Dark, before illumination . (B) 3 min illumination, (C) 4 min in the dark following illumination as in (B) . X 16,500 .

337

illumination . Hence the parameters which are pearance . Although the organization of the being measured by light scattering and transmis- chloroplasts seems unchanged by photoshrinkage, sion changes are not identical . Various theoretical a tighter ordering of the membranes in the grana (20, 21) and experimental (14, 21, 23) studies, region occurs . These samples were subjected, as especially from Shibata's laboratory (22, 36), sug- shown in Fig . 6, to analysis by microdensitometry gested that light scattering changes appear to which revealed that a decreased thickness of the measure not only volume changes, but also micro- membranes had occurred upon illumination . This structural changes occurring within the membrane surprising finding led us to undertake a more de- itself. tailed study . The changes which occur in grana membranes Electron Microscopy and Microdensitometry under different light conditions are shown in Fig . 6. The microdensitometric tracings clearly reveal The ultrastructural changes observed upon il- the reversible decrease in the spacing of 32% and luminating chloroplasts in sodium chloride-PMA thickness of 21 % upon illumination, when the medium are shown in Fig . 4. Chloroplasts are dimensions are taken as 100% in the dark . devoid of outer membranes but have a relatively Downloaded from http://rupress.org/jcb/article-pdf/47/2/332/1069424/332.pdf by guest on 29 September 2021 The presence of repetitive structures in grana intact inner membrane system where grana mem- affords a means to obtain accurate information branes are relatively loosely stacked in the dark on dimensions if the profile of the membranes on (Fig. 4 A) . After illumination in the presence of an electron micrograph is in very sharp focus PMA, these membranes become very tightly throughout the entire grana region . When, how- . 4 B), making it quite difficult packed together (Fig ever, the dimensions of single intergranal mem- to discern clear space between individual membranes are considered, the exact orientation of . Cessation of illumina- branes at low magnification membranes within the section are more difficult to tion in the presence of PMA results in a restoration ascertain . Therefore, numerous photodensitomet- of the characteristic appearance of dark chloro- ric traces of single thylakoid membranes in the plasts . These changes are seen more clearly when the results are subjected to the scrutiny of higher intergrana region were made, one of which is il- resolution electron microscopy as shown in Fig. 5, lustrated in Fig. 7. Precise values of the dimensions where grana regions are shown in a typical ap- are purposely omitted here because of the difficulty

FIGURE 5 Light-dependent reversible ultrastructural changes of chloroplast membrane system, in a sodium chloride-PMA medium. Other conditions as in Methods . (A) Dark, before illumination ; (B) 3 min illumination ; (C) 4 min following illumnination as in (B). X 46,800 .

3 3 8 THE JOURNAL OF CELL BIOLOGY • VOLUME 47, 1970 Downloaded from http://rupress.org/jcb/article-pdf/47/2/332/1069424/332.pdf by guest on 29 September 2021

FIGURE 6 Dimensions of the grana membranes in a sodium chloride-PMA medium under the different light conditions. Other conditions as in Fig. 5 and Methods . in assigning these accurately . The data, neverthe- which it was brought about, could cause similar less, indicate that all thylakoid membranes within changes? Chloroplasts suspended in the presence of the chloroplasts probably undergo the same weak acid anion solutions like sodium acetate and changes upon illumination . illuminated (as in Fig. 1) show the spacing and It was next decided to investigate whether the thickness changes as shown in Fig . 9. The thickness kinetics of light scattering and transmission decreased by 13 0/0 and the spacing changes by changes could be correlated with the thickness 27% when the chloroplasts were illuminated (Fig . and spacing changes. The results are presented in 10) . Fig. 8. These studies were made possible by the PH TITRATION STUDIES : Examination of use of the glutaraldehyde technique (6) to rapidly the ultrastructural changes at different pH's was stop changes in structure during the actual course of interest since it is known that light scattering of of the experiment . The thickness and spacing of chloroplasts is enhanced by lowering the pH of the the grana membranes, using microdensitometry medium (2, 4, 25) . When chloroplasts in 100 mm and measurements on enlarged photographs, gave NaCl solution were titrated in the dark with 0 .05 similar results . Hence within the limits of error of N HCl, light scattering increased on lowering the the measurements, thickness and spacing changes pH below 6, reaching a maximal level at pH 4, follow closely upon the corresponding photometric and then it decreased abruptly beyond pH 3 .8. tracings. The results further show that the thick- Light scattering increment induced by acidifica- ness changes more nearly correlate with the kine- tion of the suspension was reversed by realkalini- tics of light scattering responses and that the zation with 0.05 N NaOH, though the backtitra- spacing changes are more nearly reflected by tion curve did not exactly follow the forward-acid kinetics of transmission changes. titration curve and shifted slightly towards the The question was now asked whether shrinkage higher pH side. Samples for electron microscopy of chloroplast, regardless of the mechanism by were taken from the suspension at pH 7 .7 (before

SATORU MURAKAMI AND LESTER PACKER Protonation and Chloroplast Membrane Structure 339

acidification), pH 4.7 (acidification, recorded slightly hypotonie condition of the medium. Acidi- 80% light scattering increment), and pH 7 .3 (after fication to pH 4 .7 and realkalinization to pH 7 .3 realkalinization) . Chloroplasts at pH 7 .7 exhibit, did not induce notable changes in gross mor- in cross-section, swollen configuration due to the phology . Similar results of gross morphology of the chloroplasts in acid medium was reported by Dark Deamer, Crofts, and Packer (4) . Striking changes were also found with grana at higher magnification as shown in Fig . 11 . Acidifi- fication from pH 7 .7 to 4 .7 results in a decrease in the thickness and spacing of the grana membranes, which is reversed by back titration of the pH from 4 .7 to 7 .3. Fig. 12 shows the corresponding microdensitometry traces which verify that the thickness and spacing changes at different pH Light values selected for comparison of structure are the Downloaded from http://rupress.org/jcb/article-pdf/47/2/332/1069424/332.pdf by guest on 29 September 2021 same as those observed upon illumination . De- creases in thickness and spacing induced upon acidification are, respectively, 23 and 31 0/0. The pH differential chosen for these experiments is more or less equivalent to that calculated as the Dark magnitude of the light-induced proton uptake by (After illumination) chloroplasts, i.e., external pH was adjusted to simulate the magnitude of the light-induced pH gradient that would be established under conditions where chloroplasts are suspended in lightly FIGURE 7 Photodensitometric traces of a single thylakoid membrane at the intergrana region of Chloro- buffered media in the presence of strongly dis- plasts in a sodium chloride-PMA medium . sociated ions .

(%) 200 TRANSMISSION 2OiiM PMA (%)

150 100 QQ 80

60 100 (A) 44 (A) 150 240 140 200 130 120 160 110 120 100 1mn THICKNESS SPACING FIGURE 8 Comparison of kinetics of photometric and ultrastructural changes of chloroplast membrane upon illumination. Chloroplasts were incubated in 50 mm Tris-HC1, pH 8.0, containing 175 mm NaCl and 15 µn2 PMS and, as indicated, 90 µM PMA was added to the incubation . At the points indicated by verti- cal lines chloroplasts were fixed with 0 .1 M glutaraldehyde.

340 THE JOURNAL OF CELL BIOLOGY • VOLUME 47, 1970

Downloaded from http://rupress.org/jcb/article-pdf/47/2/332/1069424/332.pdf by guest on 29 September 2021

FIGURE 9 Ultrastructure of chloroplast membrane system in a sodium acetate medium under different light conditions . (A) Chloroplasts in the dark, before illumination . Both grana and intergrana thylakoids exhibit a slightly swollen configuration. (B) Chloroplasts illuminated for 3 min . Flattening of the thylakoids and contraction of thylakoid membranes have occurred . X 62,800.

SATORU MURAKAMI AND LESTER PACKER Protonation and Chloroplast Membrane Structure 34 1

reversible shrinkage of chloroplast result in both spacing and thickness changes, with the singular exception of osmotic changes which lead only to spacing changes but not to membrane thickness changes .

Studies on Refractive Index

The observed changes in membrane thickness upon illumination are indicative of changes in internal structure . Related studies were undertaken to verify if this view was correct because it Light was possible that the differences observed by microdensitometric analysis of electron micro- graphs were brought about by peculiarities in the Downloaded from http://rupress.org/jcb/article-pdf/47/2/332/1069424/332.pdf by guest on 29 September 2021 way in which chemical fixation affected structure 112+10A 144+3A of the membranes. FIGURE 10 Comparison of dimensions of grana thy- Chloroplasts were incubated in the sodium lakoid membranes in sodium acetate medium under chloride-PMA system and fixed with glutaralde- different light conditions . Conditions as in Fig . 9 and hyde before, during, and following illumination . Methods. The samples were then washed twice and resuspended in sucrose solutions of differing refractive OSMOTIC STUDIES Shrinkage of chloroplasts index (24) . The light scattering level (as a param- can also be brought about by increasing the ex- eter of refractive index difference between the ternal osmotic pressure of the medium in which chloroplast membrane and the medium) was they are suspended . To test this effect, chloroplasts plotted against the refractive index of the sucrose were suspended in a hypotonic (0 .05 M) or hyper- solution (Fig. 16) . Over a wide range of refractive tonic (0 .5M) sucrose solution containing sodium index (1 .333-1 .485), the light scattering level was chloride . Fig . 13 shows that increasing the con- always higher in illuminated chloroplasts than in centration of sucrose in a suspension leads the those in the dark . chloroplasts to contract, which results in the characteristic flattening of the membrane system . Membrane Protonation Studies This effect was caused by the osmotic gradient Since it has been disclosed by our observations corresponding to a 10-fold change in the concentration of sucrose, which was about equivalent to that pH titration can lead to changes in micro- the conditions chosen above for examining the structure, this effect was studied further . Fig. 17 transition between darkness and illumination . shows that as the pH is lowered the light scattering Fig. 14 shows the ultrastructural changes in greater increases, reaching a maximum (100-120% incre- .0 detail. A quantitative analysis of the dimension of ment) around pH 3 .8-4 . Further lowering of the pH results in a decrease of the light scattering thylakoid membranes under the different osmotic . It conditions was provided by microdensitometry is known from earlier studies by Deamer et al . (4) (Fig . 15) . It is evident that the expected spacing and Mukohata, Mitsudo and Isemura (25) that changes (32 %) are observed, but that no statisti- pH titration causes enhancement of chloroplast cally significant changes occur in thickness of the light scattering and that this change is reversible by membrane under these conditions . back titration. pH-dependent light scattering is Table III, based on statistical analysis of micro- associated with changes in internal membrane densitometric tracings, summarizes the changes of structure . Fig. 17 also shows that protonation oc- dimension in the thickness and spacing of thylakoid curs during acidification since the membrane membrane under various light, pH, and osmotic system has considerable buffer capacity in the pH conditions employed to bring about reversible range between 7 and 4.5. The amount of hydrogen shrinkage of chloroplasts . It is evident from the ions taken up was 4 .5 µeq/mg chlorophyll ; this data that all conditions employed to bring about value compares favorably with earlier reports (4) .

342 THE JOURNAL OF CELL BIOLOGY • VOLUME 47, 1970

Downloaded from http://rupress.org/jcb/article-pdf/47/2/332/1069424/332.pdf by guest on 29 September 2021

FIGURE 11 Effect of pH on ultrastructure of chloroplast membrane . Chloroplasts were incubated in a 100 mm sodium chloride solution at different pH's in the dark in a weakly alkaline medium . (A) pH 7 .7 . Before acidification ; the thylakoid membrane system exhibits swollen configuration due to slightly hypotonic condition. (B) pH 4 .7 . Upon acidification, flattening of grana thylakoids and contraction (decrease in thickness) of thylakoid membranes are induced . (C) pH 7 .3 . Reversal after medium is back-titrated with alkali. X 67,600.

SATORU MURAiAMI AND LESTER PAcIcER Protonation and Chloroplast Membrane Structure 3 43

binding sites or occupancy of ANS in the membrane as a consequence of illumination . Light-induced ANS fluorescence increases in illuminated chloroplasts in the presence of weak acid anions (sodium acetate) or strongly dissoci- ated ions (NaCI-PMA) show similar kinetics as shown in Fig . 19. In both systems the speed of the ANS fluorescence changes upon illumination are faster than the corresponding light scattering changes . Furthermore, the rate of the ANS fluorescence responses are similar in the two conditions, but the light scattering kinetics are dissimilar . The half times for ANS fluorescence changes taken from these first order plots indicate that they are 14 sec in the NaCl-PMA system and 15 sec in Downloaded from http://rupress.org/jcb/article-pdf/47/2/332/1069424/332.pdf by guest on 29 September 2021 sodium acetate system . These rates are consider- ably faster than light scattering changes, where the half time for changes are 60 and 33 sec for the sodium chloride-PMA and sodium acetate systems, respectively . pH 7 .3 Since kinetic analysis of the ANS fluorescence changes indicated that they occurred more rapidly V than the light scattering responses, it was of considerable interest to determine how ANS fluorescence changes are related to structural changes in thylakoid membrane . Since it is known that il- FIGURE 12 pH dependence of dimensions of grana lumination of chloroplasts leads to an uptake of membranes. Chloroplasts were incubated in 100 mm hydrogen ions, the effect of protonation of chloro- sodium chloride at the pH indicated in the dark. plasts on ANS fluorescence was studied . Fig . 20 shows experiments in which pH titrations of the light scattering and ANS fluorescence changes are ANS Fluorescence Studies compared . It is clear that the fluorescence changes ANS is known to enhance its ultraviolet-excited induced by protonation of the membrane are more fluorescence when it becomes associated with sensitive to pH change than light scattering. When hydrophobic environments such as occur when it a similar titration was made with potassium (as is bound to biological membranes or placed in chloride), no increases in either ANS fluorescence solutions of low dielectric constant (26) . When or light scattering are observed . Similar negative ANS is added to the chloroplasts, fluorescence is results have been obtained with other monovalent enhanced by accompanying binding to the mem- ions. brane (Fig . 18) . Upon illumination a further increase in this fluorescence intensity is observed . DISCUSSION Chloroplasts treated with glutaraldehyde are no longer capable of undergoing changes in volume From the findings in this investigation, it may be as measured by light scattering, transmission, or concluded that the temporal sequence of events electron microscopy. Nevertheless, it was found involved in the hierarchy of changes in structure that the light-dependent increases in the ANS in chloroplast membranes following illumination fluorescence still occurs . This finding suggests that probably is : (a) protonation, (b) change in the changes in microstructure of the membrane resist environment within the membrane, (c) change in chemical fixation sufficient to prevent gross membrane thickness, (d) change in internal osmo- morphological changes . The difference in ANS larity accompanying ion movements with conse- fluorescence observed between fixed and unfixed quent collapse and flattening of thylakoids, (e) chloroplasts probably represents changes in change in the gross morphology of the inner

344 THE JOURNAL OF CELL BIOLOGY • VOLUME 47, 1970

Downloaded from http://rupress.org/jcb/article-pdf/47/2/332/1069424/332.pdf by guest on 29 September 2021

FIGURE 13 Ultrastructure of osmotically contracted chloroplasts . (A) Hypotonie (0 .05 M) sucrose-sodium chloride medium in the dark . Chloroplasts appear round, and grana are easily resolved. (B) Hypertonic (0.5 M) sucrose-sodium chloride medium in the dark . Osmotically-induced flattening of the membrane system is evident . X 8800 .

SATORU MURAKAMI AND LESTER PACKER Protonation and Chloroplast Membrane Structure 3 4 5 Downloaded from http://rupress.org/jcb/article-pdf/47/2/332/1069424/332.pdf by guest on 29 September 2021

FIGURE 14 Ultrastructure of chloroplast membranes under hypotonie and hypertonic conditions . (A) 0.05 M sucrose-sodium chloride medium in the dark. (B) 0 .5 sucrose-sodium chloride medium in the dark. X 67,200.

chloroplast membrane system, and (f) change in the gross morphology of intact chloroplasts . 0.05 M A mechanism which relates membrane "proto- Sucrose-NaCl nation" to this hierarchy of changes in structure in chloroplast membrane is presented in Fig . 21 . The evidence in support of this hierarchy of changes in structure is presented below.

Protonation and Microstructural Changes The evidence in support of the idea that proto- 0 .5 M nation is involved in the decrease in thickness can Sucrose-NaCl be summarized as follows : (a) Several treatments (light, H+, hypertonicity) have been found to decrease membrane spacing ; all of them cause a decrease in membrane thickness, except an osmotic force . Those treatments, again except for osmotic changes, have in common a protonation resulting FIGURE 15 Effect of osmolarity upon dimensions of grana thylakoid membrane . Chloroplasts were incu- from illumination or acidification in the dark . (b) bated in either 0 .05 M or 0 .5 M sucrose solution con- By using artificial means (titration with HCI), in taining 44 mm NaCl and 12 .5 mM Tris -HCI, pH 7 .9 the dark the acidification of the membrane causes in the dark . an increase in light scattering and a decrease in

3 4 6 THE JOURNAL OF CELL BIOLOGY • VOLUME 47, 1970

TABLE III

Effect of Incubation Conditions upon Dimensions of the Grana Thylakoid Membrane

Condition Thickness (so) Spacing (so)

(A) (A) Light-induced shrinkage Dark -, light --o dark Dark --> light --o dark NaCl-PMA 131 f 10 104±6 130±9 212 f8 144±9 214±4 Sodium acetate 129 ± 9 112 ± 10 196 ± 4 144 ± 3

pH-induced shrinkage (dark) pH 7.7 -o 4 .7 -o 7 .3 pH 7 .7 ---o 4 .7 ---+ 7.3 NaCl 136 ± 8 105 ± 7 138 ± 11 206 ± 15 143 ± 6 236 ± 24

Osmotic shrinkage 0 .05 M sucrose-NaCl 130 =L 8 252 ± 26 0 .5 m sucrose-NaCl 125 ± 7 172 ± 5 Downloaded from http://rupress.org/jcb/article-pdf/47/2/332/1069424/332.pdf by guest on 29 September 2021

150 O

s °-' 50 0 O o •

0 1 1_ 1 I 1 1 1 .32 1 .36 1 .40 1 .44 1 .48 Refractive index of medium FIGURE 16 Influence of refractive index of the suspending medium upon 90 ° light scattering of chloroplast membranes. Chloroplasts were incubated in a sodium chloride-PMA medium in the dark, light (4 min) or dark (10 min) after 4 min illumination, and then fixed with 0 .1 M glutaraldehyde and washed thoroughly by centrifugation . The fixed chloroplasts, containing 10 4ug chlorophyll/ml, were suspended in 3.5 ml sucrose solution of different concentrations (0-85%). The light scattering level of the chloroplasts, which were fixed in the dark and suspended in distilled water, was adjusted at 100% . Refractive index of suspending medium was estimated from refractive index of aqueous solution of sucrose (23) . Chloroplasts in the dark (before illumination) (D),- •- ; light (4 min) (L),-0- ; dark (10 min) after 4 min illumination (L-D), -A-.

the thickness of the membrane in a reversible changes can be satisfactorily accounted for by a fashion if back titration is instituted . Changes in prior protonation. internal structure of the membranes monitored by Three considerations suggest that a study of light scattering are correlated closely with the fixed charges on the chloroplast membrane war- protonation of the membranes between pH 7 .0 and rants further investigation . 4.5. (c) Changes in ANS fluorescence by chloro- The first is that light-induced ANS fluorescence plast membranes have been shown to occur at a intensity changes are retained in glutaraldehyde- faster rate than light scattering changes and to be fixed chloroplasts. Hence, changes in microstruc- pH dependent. In both of these instances the ture of the membrane can occur independent of

SATORU MURAKAMI AND LESTER PACKER Protonation and Chloroplast Membrane Structure 347

20MM ô PMA Unfixed 200 chloroplast ô 120 Q) a) a u Fixed h 150 a) 100 - chloroplast S rn 0 ON J i 100 s• o V) o. Z Q 0 _

i 1 min 100µg 2 a. ANS

FIGURE 18 Light-induced ANS fluorescence changes

of chloroplasts . 100 µg ANS was added to the chloro- Downloaded from http://rupress.org/jcb/article-pdf/47/2/332/1069424/332.pdf by guest on 29 September 2021 C plasts equivalent to 14 µg chlorophyll per ml in 50 mat Tris-HCI, pH 8.0, containing 175 mart NaCl and 15 Am 1 .0 - PMS. When fluorescence yield in the dark reached the maximal level (100%), 20 )um PMA was added and then chloroplasts were illuminated by red light . For the experiment with fixed chloroplasts, chloroplasts were fixed with 0 .1 as glutaraldehyde in the dark, thoroughly I I I I I I washed, and incubated. 1 2 3 4 5 6 7 H* added (µeq)

FIGURE 17 Comparison of 90 ° light scattering with pH and protonation of the chloroplast membrane . Chloroplasts containing 200 µg chlorophyll were suspended in 7 .5 ml of 100 mm sodium chloride and titrated with 0.05 N HCl in the dark . (A) 90 ° light scattering . (B) pH changes. (C) protonation . Hydrogen ion concentrations were calculated from pH values measured for both chloroplast suspension and suspending medium, and the difference between these two values was taken as the amount of proton uptake .

changes in gross morphology and configuration, which do not occur under these conditions of fixation. Possible explanations for ANS fluorescence changes are : (a) Protonation of negative charges in the membrane would lead to an expulsion of attractive water dipoles localized in the vicinity of the negatively charged groups . This would also have the effect of increasing the attractive forces between the molecules, resulting in a decrease in 90° light scatter- membrane thickness as a consequence of the closer FIGURE 19 Kinetics of light-induced fluorescence changes in chloroplasts . . ing and ANS apposition of unit structures within the membrane Chloroplasts equivalent to 14 µg chlorophyll per ml This type of effect would also increase the "hydro- were suspended in either a sodium chloride-PMA or phobicity" of the membrane. (b) Increase in the sodium acetate medium . Light scattering and ANS hydrophobicity of the thylakoid membrane might fluorescence increase upon illumination were expressed be brought about by a combination of effects, as the difference between the value at steady state and among which are polarization of fixed negative that at various times on a semilog scale .

348 THE JOURNAL OF CELL BIOLOGY • VOLUME 47, 1970 300 pation of these negatively charged groups in protonation of the membrane seems to be plausible from the fact that aspartic and glutamic acids, which have side carboxyl groups, are the most predominant amino acids in the protein of the chloroplast membrane (32, 33) . 250 250 The third reason is that charge changes of the membrane as a result of protonation would alter the attractive and repulsive forces causing changes ô ô in quaternary protein structure . These could lead 0) to a 20% decrease in membrane dimensions ob- C 200 d served . Cassim and Yang (34) examined the pH â dependence of conformation of poly-L-glutamic Nu acid by circular dichroism and optical rotatory dispersion techniques . Their results suggest that V)

so! Downloaded from http://rupress.org/jcb/article-pdf/47/2/332/1069424/332.pdf by guest on 29 September 2021 Z J protonation of carboxyl groups causes aggregation Q ô in acid solution . Also, Sun (35) has shown that the 150 150 conformation of bovine serum albumin is a func- tion of electrostatic charge of the molecule and, hence, in this case it exists in a compact form between pH 4 .3 and 10.5 where negatively charged groups are protonated.

100 100 Configurational and Gross Morphological Changes HCI or KCI added (µmoles) Illumination of chloroplasts in the presence of FtcuRE 20 Effect of H+ and K+ on 90 ° light scatter- weak acid anions (5, 27, 28) or the NaCl-PMA ing and ANS fluorescence intensity of chloroplasts . system (14, 18, 19) results in their flattening or Spinach chloroplasts were incubated in 100 mnz sodium decrease of volume . This change decreases the chloride, and 90 ° light scattering and ANS fluorescence spacing between grana membranes, as is particu- were titrated with 0 .05 N HO or 0 .05 x KC1 as de- larly clearly revealed byamicrodensitometry studies scribed in Methods. reported herein and by related low angle X-ray scattering studies . Comparison of the kinetics of charges, expulsion of water, and neutralization of changes in membrane structure and photometric the phospholipids . parameters (Figs . 3, 8) further shows that these The second reason is that pH titration studies spacing changes are best correlated with changes suggest that protonation is affecting only certain in transmission, which is known to most closely fixed charges, such as carboxyl groups, on the correlate with volume changes . Therefore, it is membrane . In support of this, it is found that the reasonable to assume an osmotic basis for the buffer capacity of the membrane between pH 4 .5 observed spacing changes . Hence, the initial and 7 is roughly equivalent to 4 .5 µeq protons per osmolarity of the medium in which chloroplasts mg of chlorophyll . A similar value was reported by are suspended would be expected to play a signifi- Deamer et al . (3) . This value for the buffer capa- cant role in determining the magnitude of spacing city of the membrane would most nearly correlate changes . Deamer and Packer (29), by employing with the neutralization of ß and y carboxyl and a special rapid centrifugation technique, have imidazole groups of protein and secondary phos- demonstrated by isotopic analysis some quantita- phate groups of lipids as being the most likely tive and qualitative aspects of anion transport that negative charges neutralized by the protonation are commensurate with explaining volume (size) process, since their pK values (30, 31) are in the and configuration (shape) changes upon an osmo- pH range at which protonation occurred . Partici- tic basis .

SATORIJ MURAKAMt AND LESTER PACKER Protonation and Chloroplast Membrane Structure 349

THYLAKOID MEMBRANE METHODS

(Thickness) (Spacing)

200-210A EM] - '/2)130 .135) A I- C 180A X-ray

Light

Dark

Dark Alkalinization

H 2 O" R _ ,-,,, Downloaded from http://rupress.org/jcb/article-pdf/47/2/332/1069424/332.pdf by guest on 29 September 2021 Light - Acidification H +

Changes in conformation Dehydration -> H 2 O and interaction of molecules ANS fluor . C

Increase of (INSIDE) by drophobicity (OUTSIDE)

1 Decrease of membrane thickness and increase of refractive index

--- %2(100-110)A > 140 -145A [ EM] (Thickness) 167A [X-ray d iffr . (Spacing,

FIGURE 21 Proposed mechanism for structural changes in chloroplast membrane by protonation .

The authors are grateful to Miss Susan Tinsley for dark (expanded) and in illuminated (contracted) expert technical assistance with the electron micros- conditions the spacing changes as observed by low copy, facilities for which were generously provided angle X-ray scattering. Preliminary results indicate in the laboratories of the Veterans Administration that a change from 180 to 167 A occurs after il- Hospital, Martinez, California, and to Professor lumination of chloroplasts in a sodium chloride- Daniel Branton for use of his microdensitometry PMA medium . apparatus. Dr. Murakami is on leave from the Department of In collaboration with Doctors V . Luzzati and T. Biology, University of Tokyo, Komaba, Tokyo . Gulik-Krzywicki (unpublished results), we have This research was supported by the National studied in glutaraldehyde-fixed chloroplasts in the Science Foundation (GB-7541) .

35 0 THE JOURNAL OF CELL BIOLOGY . VOLUME 47, 1970

Received for publication 6 November 1969, and in revised 18 . SIEGENTHALER, P. A ., and L . PACKER . 1965 . form 27 January, 1970. Plant Physiol . 40 :785 . 19 . SIEGENTHALER, P. A. 1966 . Physiol. Plant . 19 :437 . REFERENCES 20. LATIMER, P., D . M . MOORE, and F. D . BRYANT . 1968 . J. Theor . Biol. 21 :348 . 1 . NEUMANN, J., and A. T. JAGENDORF . 1964 . 21 . BRYANT, F. D ., P . LATIMER, and B. A. SEIBER . Arch . Biochem . Biophys. 107 :109 . 1969 . Arch . Biochem . Biophys . 135 :109. 2 . DILLEY, R. A ., and L . P. VERNON . 1965 . Arch. 22 . YAMASHITA, K., M. ITOH, and K. SHIBATA . Biochem. Biophys . 111 :365 . 1968 . Biochim . Biophys . Acta . 162 :610. 3 . DILLEY, R. A., R . B . PARK, and D. BRANTON . 23 . MUKOHATA, Y. 1966 . Annu . Rep . Biol. Works, 1967 . Photochem. Photobiol . 6 :407 . Fac . Sci. Osaka Univ . 14 :121 . 4 . DEAMER, D. W., A. R. CROFTS, and L. PACKER . 24 . WEAST, R. C ., and S. M . SELBY. 1966 . In Hand- 1967. Biochim . Biophys . Acta . 131 :81 . book of Chemistry and Physics . Chemical 5 . CROFTS, A. R., D . W . DEAMER, and L. PACKER . Rubber Company Cleveland, Oh . E 154. 1967. Biochim . Biophys . Acta . 131 :97. 25 . MUKOHATA, Y., M. MITSUDO, and T. ISEMURA . 6. DEAMER, D. W., and L. PACKER . 1967 . Arch . 1966 . Annu. Rep . Biol . Works, Fac. Sci ., Osaka

Biochem . Biophys. 119 :83 . Univ . 14 :107 . Downloaded from http://rupress.org/jcb/article-pdf/47/2/332/1069424/332.pdf by guest on 29 September 2021 7 . SHAVIT, N., and M . AVRON . 1967. Biochim . 26 . STRYER, L . 1965 . J. Mol. Biol. 13 :482 . Biophys . Acta . 131 :516. 27 . PACKER, L. 1963 . Biochim . Biophys. Acta . 75 :12. 8 . KARISH, S . J . D ., and M. AVRON . 1968 . Biochim . 28 . ITOH, M., S . IZAWA, and K. SHIBATA . 1963 . Biophys . Acta. 153 :878 . Biochim . Biophys. Acta . 66 :319 . 9 . PACKER, L., J . M. ALLEN, and M . STARKS . 29. DEAMER, D. W ., and L . PACKER . 1969 . Biochim . 1968 . Arch . Biochem . Biophys. 128 :142 . Biophys. Acta . 172 :539. 10 . PACKER, L., A . C. BARNARD, and D. W. DEAMER . 30. TANFORD, C. 1962 . Advan . Protein Chem . 17 :69 . 1967 . Plant Physiol. 42 :283 . 31 . BANGHAM, A. D . 1963 . Advan. Lipid Res. 1 :65 . 11 . NOBEL, P . S . 1968 . Plant Cell Physiol . 9 :499. 32 . LOCKSHIN, A., and R . H . BURRIS . 1966. Proc. 12 . NOBEL, P . S . 1969 . Biochim . Biophys . Acta . 172 :134. Nat. Acad. Sci. U.S.A . 56 :1564. 13 . MURAKAMI, S., and L Plant . PACKER . 1970 . 33 . CRIDDLE, R . S . 1966. In Biochemistry of Chloro- . Physiol 45 : 289 . plasts. T. W. Goodwin, editor, Academic 14 . MURAKAMI, S., and L . PACKER . 1969 . Biochim . Press Inc ., New York. 1 :203 . Biophys. Acta . 180 :420. 34 . CASSIM, J. Y., and J . T. YANG . 1967 . Biochem . 15 . MILLER, V. L ., D. POLLEY, and C. J. GOULD . 1951 . Anal . Chem. 23 :1286 . Biophys . Res. Commun. 26 :58. 16. MOLLENHAUER, H. H . 1964. Stain Technol. 39 :111 . 35 . SUN, S . F . 1969 . Arch . Biochem . Biophys. 129 :411 . 17 . JOHNSON, U. G., and K . R. PORTER. 1968 . J. 36 . ITCH, M., S . IZAWA, and K. SHIBATA . 1963 . Cell Biol . 38 :403 . Biochim . Biophys . Acta . 69 :130 .

SATORU MURAKAMI AND LESTER PACKER Protonation and Chloroplast Membrane Structure 35 1